Abstract
The basic premise of the hybrid von Neumann-dataflow or multithreaded model of computation is that exploiting locality within programs can impact performance dramatically. In this paper we discuss the nature of various forms of locality and their impact on the efficiency of execution. One form of locality: intra-thread locality is defined, quantified and experimentally evaluated using a set of benchmarks. The results point to a very large degree of thread locality: over half the instructions have to wait less than 5 instruction execution steps for their input data. Furthermore, there is a remarkable uniformity and consistency of the distribution of thread locality across a wide variety of benchmarks.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Arvind and R.A. Iannucci. Two fundamentals issues in multiprocessors: the data-flow solutions. Technical Report LCS CSG 226–6, Laboratory for Computer Science, MIT, Cambridge Massachusetts, May 1987.
A. P. W. Böhm and J. R. Gurd. Iterative instructions in the Manchester dataflow computer. IEEE Trans. on Parallel and Distributed Systems, 1(2):129–139, 1990.
Steven Brobst. Organization of an instruction scheduling and token storage unit in a tagged token dataflow machine. In S. K. Sahni, editor, Proc. of the 1987 International Conf. on Parallel Processing, pages 40–45, 1987.
W.P. Crowley, C.P. Hendrickson, and T.E. Rudy. The simple code. Technical Report TR UCID 17715, Lawrence Livermore National Laboratory, 1978.
D. E. Culler and G. M. Papadopoulos. The explicit token store. Journal of Parallel and Distributed Computing, 10(4), 1990.
J. T. Feo. The Livermore Loops in Sisal. Technical Report UCID-21159, Computing Research Group, Lawrence Livermore National Laboratory, Livermore, CA 94550, August 1987.
V. G. Grafe and J. E. Hoch. The Epsilon-2 hybrid dataflow architecture. In Compcon’90, pages 88–93, 1990.
V. G. Grafe and J. E. Hoch. The EPSILON-2 multiprocessor system. Journal of Parallel and Distributed Computing, 10(4), 1990.
H. J. Hum and G. R. Gao. A novel high-speed memory organization for fine-grain multi-thread computing. In Proc. of Parallel Architecture and Languages Europe ‘81, 1991.
J. L. Hennessy and D. A. Patterson. Computer Architecture: A Quantitative Approach. Morgan Kaufmann Publishers Inc., 1990.
R. A. Iannucci. A dataflow/von Neumann hybrid architecture. Technical Report MIT/LCS/TR-418, MIT Laboratory for Computer Science, May 1988.
F. H. McMahon. Livermore fortran kernels: A computer test of numerical performance range. Technical Report UCRL-53745, Lawrence Livermore National Laboratory, Livermore, CA, December 1986..
J. McGraw, S. Skedzielewski, S. Allan, R. Oldehoeft, J. Glauert, C. Kirkham, B. Noyce, and R. Thomas. SISAL: Streams and Iteration in a Single Assignment Language: reference manual version 1.2. Manual M-146, Rev. 1, Lawrence Livermore National Laboratory, Livermore, CA, March 1985.
W. Najjar and J.-L. Gaudiot. Multi-level execution in Data-Flow architectures. In S. K. Sahni, editor, Proc. International Conf. Parallel Processing, pages 32–39, St. Charles, Illinois, 1987.
Walid A. Najjar, W. Marcus Miller, and A. P. Wim Böhm. An analysis of loop latency in dataflow execution. In Proceedings of the 19th International Symposium on Computer Architecture, 1992.
J. R. Rice. Problems to test parallel and vector languages. Technical Report CSD-TR 516, Purdue University, 1985.
J. R. Rice. Problems to test parallel and vector languages - 2. Technical Report CSD-TR 1016, Purdue University, 1990.
S. Sakai, Y. Yamaguchi, K. Hiraki, Y. Kodama, and T. Yuba. An architecture of a dataflow single chip processor. In Int. Symp. on Computer Architecture, pages 46–53, May 1989.
S. Sakai, Y. Yamaguchi, K. Hiraki, Y. Kodama, and T. Yuba. Pipeline optimization of a dataflow machine. In J-L. Gaudiot and L. Bic, editors, Advanced Topics in Data-Flow Computing, pages 225–246. Prentice Hall 1991.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1994 Springer Science+Business Media New York
About this chapter
Cite this chapter
Najjar, W.A., Böhm, A.P.W., Miller, W.M. (1994). Exploiting Locality in Hybrid Dataflow Programs. In: Iannucci, R.A., Gao, G.R., Halstead, R.H., Smith, B. (eds) Multithreaded Computer Architecture: A Summary of the State of the ART. The Springer International Series in Engineering and Computer Science, vol 281. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-2698-8_15
Download citation
DOI: https://doi.org/10.1007/978-1-4615-2698-8_15
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-6161-9
Online ISBN: 978-1-4615-2698-8
eBook Packages: Springer Book Archive